Porphyromonas gingivalis is the major etiological agent of adult periodontitis. Although well established as a periodontal pathogen, it has more recently been recognized that P. gingivalis may have an important role in systemic disease, including cardiovascular diseases, prematurity and low birth weight. Cardiovascular diseases are a heterogeneous group of conditions that are the leading cause of death in developed countries. The economic losses and psychological repercussions caused by this pathogen are thus potentially enormous. During the past several years, we have investigated the pathogenic properties of P. gingivalis, focusing on its interactions with and invasion of human tissues, including coronary artery endothelial cells. However, the availability of the whole genome sequence of only one strain of P. gingivalis has hampered our research effort. Many factors suggest that investigation of other P. gingivalis strains should be pursued since strains are phenotypically and genotypically heterogeneous with regards to certain virulence factors including relative host cell invasive abilities. Consequently, this submission is requesting funding as a supplement to our recently funded competitive renewal (DE13545) to sequence and annotate the genomes of P. gingivalis strains 381, 33277, A7436, and AJW4, followed by comparative genomic analysis. Genome sequences will be determined using a new, cost-efficient, and massively parallel DMA sequencing technology developed by 454 Life Sciences (Branford, CT). Genome assembly will be supplemented with up to 1x genome coverage using paired-end Sanger DNA sequencing reads. Gene location and structures will be predicted using FGENESB. Gene function and names will be assigned to predicted gene structures based on homology searches of public gene data. Pairwise genome comparisons and multiple genome comparisons will be performed to identify orthologs that are conserved among these five genomes as well as unique genes. Genes that are missing from these genomes will also be determined. BLAST searches, MUMmer programs and VISTA tools will be used for these purposes. For all predicted proteins, a HMMER PFAM search will be conducted to detect functional domains. The availability of four additional whole genome sequences of P. gingivalis in public databases would facilitate basic research, including that of the parent grant to this supplemental application, and more rapid development of diagnostic tests, and treatment, and prevention strategies.